UGC-VQA: Benchmarking Blind Video Quality Assessment for User Generated Content
Open accessThe DOI matches its Crossref record; 92 of the 94 references it cites were found in a registry or an index, and 2 are not checked yet; 10 of the 27 sources its claims cite were checked for support.
Abstract
Recent years have witnessed an explosion of user-generated content (UGC) videos shared and streamed over the Internet, thanks to the evolution of affordable and reliable consumer capture devices, and the tremendous popularity of social media platforms. Accordingly, there is a great need for accurate video quality assessment (VQA) models for UGC/consumer videos to monitor, control, and optimize this vast content. Blind quality prediction of in-the-wild videos is quite challenging, since the quality degradations of UGC videos are unpredictable, complicated, and often commingled. Here we contribute to advancing the UGC-VQA problem by conducting a comprehensive evaluation of leading no-reference/blind VQA (BVQA) features and models on a fixed evaluation architecture, yielding new empirical insights on both subjective video quality studies and objective VQA model design. By employing a feature selection strategy on top of efficient BVQA models, we are able to extract 60 out of 763 statistical features used in existing methods to create a new fusion-based model, which we dub the VIDeo quality EVALuator (VIDEVAL), that effectively balances the trade-off between VQA performance and efficiency. Our experimental results show that VIDEVAL achieves state-of-the-art performance at considerably lower computational cost than other leading models. Our study protocol also defines a reliable benchmark for the UGC-VQA problem, which we believe will facilitate further research on deep learning-based VQA modeling, as well as perceptually-optimized efficient UGC video processing, transcoding, and streaming. To promote reproducible research and public evaluation, an implementation of VIDEVAL has been made available online: https://github.com/vztu/VIDEVAL.
Verified 2026-11-20 · 2 references not checked yet
Overview
CoChat Summary
TL;DR
This paper benchmarks blind video quality assessment (BVQA) models on large-scale user-generated content (UGC) video databases and proposes VIDEVAL, a feature-fusion model selected from 60 of 763 statistical features. VIDEVAL achieves state-of-the-art or near-state-of-the-art performance at lower computational cost than leading models, and the study provides a reproducible benchmark for UGC-VQA.
Key findings
- VIDEVAL selected 60 of 763 features and achieved SOTA/near-SOTA on all test sets: All-Combinedc SRCC 0.7960, PLCC 0.7939; KoNViD-1k SRCC 0.7832, PLCC 0.7803; LIVE-VQC SRCC 0.7522, PLCC 0.7514; YouTube-UGC SRCC 0.7787, PLCC 0.7733.
- TLVQM outperformed other BVQA models on LIVE-VQC by a notable margin (SRCC 0.7988, PLCC 0.8025) but underperformed on YouTube-UGC (SRCC 0.6693, PLCC 0.6590); models containing temporal features excelled on LIVE-VQC.
- Completely blind NIQE, ILNIQE, and VIIDEO performed poorly; on All-Combinedc, NIQE had SRCC 0.4622/PLCC 0.4773 and VIIDEO had SRCC 0.1039/PLCC 0.1621.
- Deep CNN baselines transferred well: ResNet-50 on KoNViD-1k had SRCC 0.8018/PLCC 0.8104; ensembling VIDEVAL with deep features improved performance by up to ∼4%, e.g., VIDEVAL+ResNet-50 on All-Combinedc had SRCC 0.8115/PLCC 0.8286.
Methods
- Evaluated leading BVQA/BIQA models on KoNViD-1k (1,200 videos), LIVE-VQC (585), YouTube-UGC (1,380; 1,323 color videos as YouTube-UGCc), and All-Combinedc (3,108) after INLSA MOS calibration; used SVR regression, 80%/20% train-test splits repeated 100 times, and reported SRCC, KRCC, PLCC, and RMSE.
- VIDEVAL was built from an initial 763-dimensional feature set drawn from BRISQUE, GM-LOG, HIGRADE-GRAD, FRIQUEE variants, and TLVQM; feature selection used random forest, linear SVM, and SFFS, choosing SVM importance with k=60 over 100 random splits; spatial features were computed every second frame within 1-second chunks with average and standard deviation pooling, then average pooled across chunks.
- Additional analyses included categorical subsets by resolution, content, and quality; cross-dataset generalization; eleven temporal pooling methods; complexity/runtime comparisons; and ensembling VIDEVAL with deep features.
Limitations
- Paper-stated: MOS standard deviation analysis is not presented because LIVE-VQC does not provide it; YouTube-UGC contains 57 grayscale videos that cause numerical errors for FRIQUEE, so color models were evaluated on a 1,323-video subset; temporal pooling results are only self-consistent and not comparable to prior experiments; not enough samples (8–20) per video were available for statistically meaningful pooling; non-natural/computer-generated content remains hard and lacks sufficiently large UGC-VQA datasets.
- Evident from the text: the benchmark covers three databases and a calibrated composite; categorical subsets for animation (81), screen content (163), and gaming (209) are small; cross-dataset generalization is modest (best SRCC values in Tables IX/X range roughly 0.33–0.70); and the deep models evaluated are mostly image-pretrained or picture-quality models rather than video-trained temporal CNNs.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
7 Valid · 1 Not valid · 9 Not checked · 69 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“These two databases, however, only modeled (camera) capture distortions on small numbers of not very diverse unique contents.”
What the sources say
- Partly supportedcites [12] [12] M. Nuutinen, T. Virtanen, M. Vaahteranoksa, T. Vuori, P. Oittinen, and J. H¨ akkinen, 'CVD2014-a database for evaluating no-reference video quality assessment algorithms,' IEEE Trans. Image Process. , vol. 25, no. 7, pp. 3073-3086, 2016. - sourceExperimental — accuracy not yet measured
Our copy supports the CVD2014 half only in part: the Abstract/Introduction and the repository Description state that CVD2014 models real camera-acquisition distortions ('the videos in the CVD2014 database are captured using 78 different cameras, and the distortions are related to the video acquisition process') on a small content base of 234 videos from five different scenes, and Section II.A notes the sequences had to be 'shot by different cameras and still be as similar as possible'. However, the paper never characterizes its own content as 'not very diverse' — Section II.A instead says the sequences 'were captured from many different scenes' — and it makes no evaluative claim of limited unique content. Nothing in our copy addresses LIVE-Qualcomm at all: Table I lists only LIVE [11] and LIVE Mobile [12], both with compression and transmission-error distortions, not capture distortions, so that half of the sentence finds no support here. (LLM, conf=0.70)
- Not checked — no full text of the sourcecites [13] [13] D. Ghadiyaram, J. Pan, A. C. Bovik, A. K. Moorthy, P. Panda, and K.-C. Yang, 'In-capture mobile video distortions: A study of subjective behavior and objective algorithms,' IEEE Trans. Circuits Syst. Video Technol. , vol. 28, no. 9, pp. 2061-2077, 2017. - source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Indeed, if temporal distortions (like judder) are present, they may severely underperform if the frame quality is high [81].”
What the sources say
- Supportedcites [81] [81] P. C. Madhusudana, X. Yu, N. Birkbeck, Y. Wang, B. Adsumilli, and A. C. Bovik, 'Subjective and objective quality assessment of high frame rate videos,' arXiv preprint arXiv:2007.11634 , 2020. - source
The Table 6 discussion says that when high spatial quality is combined with low temporal quality that a model does not assess, spatial measurements unduly influence its video-quality prediction. It also reports that VMAF fails to capture judder/stutter artifacts. This supports the claim for models that inadequately account for temporal distortions. (LLM, conf=0.91)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“The most probable reason would be that these models were trained on picture quality datasets [64], [70], which contain different types of (strictly spatial) distortions than UGC-VQA databases.”
What the sources say
- Partly supportedcites [64] [64] Z. Ying, H. Niu, P. Gupta, D. Mahajan, D. Ghadiyaram, and A. Bovik, 'From patches to pictures (PaQ-2-PiQ): Mapping the perceptual space of picture quality,' in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR) , 2020. - sourceExperimental — accuracy not yet measured
Our copy supports the identification half of the attribution: PaQ-2-PiQ's database is a still-picture quality database whose contents are impaired by 'complex mixtures of highly diverse distortions' (Table 1 caption; Sec. 2 'Image Quality Datasets'; Sec. 3.1 UGC-like picture sampling). What the source does not support is the comparative/explanatory part of the claim: nowhere does our copy contrast these distortion types with those of UGC video (VQA) databases, characterise its distortions as 'strictly spatial' in opposition to temporal video distortions, or offer any account of why a picture-quality-trained model performs poorly on video. Searches for video/temporal-distortion comparisons returned only passing mentions (streaming services, VMAF, 'pictures and videos' in Sec. 5). (LLM, conf=0.70)
- Supportedcites [70] [70] V. Hosu, H. Lin, T. Sziranyi, and D. Saupe, 'Koniq-10k: An ecologically valid database for deep learning of blind image quality assessment,' IEEE Trans. Image Process. , vol. 29, pp. 4041-4056, 2020. - source
The citation identifies a picture-quality dataset, and the Introduction (p. 1) identifies KonIQ-10k as a database of images with quality ratings. The Subjective Image Quality Assessment section (p. 4) lists the image degradations considered in those ratings. The proposed reason for the models’ poor performance is the citing authors’ inference, not a finding attributed to this source. (LLM, conf=0.90)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The empirical results show that a simple aggregation of these known models can achieve state-of-the-art (SOTA) performance.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Performance results are shown in Table VIII, wherein VIDEVAL still outperformed the other BVQA models on both low and high quality partitions.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
43 more claims
Create a free account to see all 86 claims and to verify the 6 references still open.
- Probably not a substantial claim. 24
- It may state two separate claims as one. 20
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ F. Complexity Analysis and Runtime Comparison · p. 13
“Deep ConvNet features ranked second in dimension (VGG-19 (4,080) and ResNet-50 (2,048)).”
§ A. Evaluation Protocol · p. 8
“The fully-connected layer (4,096-dim) from VGG-19 and average-pooled layer (2,048-dim) from ResNet50 served as deep feature descriptors”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ C. Performance Evaluation on Categorical Subsets · p. 11
“The proposed VIDEVAL model outperformed over all categories, followed by ResNet-50 and FRIQUEE, suggesting that VIDEVAL features are robust quality indicatives across different content categories.”
§ B. Performance on Individual and Combined Datasets · p. 10
“| ResNet-50 | 0.6199 | 0.6676 | 0.2781 | 0.4871 | 0.6378 | 0.6779 |”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“| Number of ratings | 136,800 (114 votes/video) | 205,000 (240 votes/video)”
Reported 240 · recomputed 350.42735
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ UGC-VQA: Benchmarking Blind Video Quality Assessment for User Generated Content · p. 1
“Our experimental results show that VIDEVAL achieves state-of-theart performance at considerably lower computational cost than other leading models.”
§ F. Complexity Analysis and Runtime Comparison · p. 13
“| TLVQM | MATLAB-CPU | 183.8 | | VIDEVAL | MATLAB-CPU | 305.8 |”
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333 · cited 3×
“To conquer the limits of data scale, they either propose to conduct patch-wise training [62], [63], [66] using global scores, or by pretraining deep nets on ImageNet [67], then fine tuning.”
MethodImage quality assessment: from error visibility to structural similarity
Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004) · IEEE Transactions on Image Processing · doi:10.1109/tip.2003.819861 · Cited by 57,892 · cited 2×
“Some popular FR-VQA models, including PSNR, SSIM [25], and VMAF [21] have already been successfully and massively deployed to optimize streaming and shared/uploaded video encoding protocols by leading video service providers.”
MethodSelf-citationLIBSVM: A library for support vector machines
Chang, C.-C., & Lin, C.-J. (2011) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/1961189.1961199 · Cited by 40,918 · cited 1×
“As recommended in [80], we optimized the SVR parameter values (C, γ ) by a grid-search of 10×10 exponentially growing sequences (in our experiments, we used a grid of C = 21, 22, .”
MethodMaking a “Completely Blind” Image Quality Analyzer
Mittal, A., Soundararajan, R., & Bovik, A. C. (2013) · IEEE Signal Processing Letters · doi:10.1109/lsp.2012.2227726 · Cited by 6,766 · cited 8×
“before (Figure 10a) and after (Figure 10b) INLSA linear mapping, calibrated by NIQE [53] scores.”
MethodInfluentialSelf-citationMultiscale structural similarity for image quality assessment
Wang, Z., Simoncelli, E. P., & Bovik, A. C. (2004) · The Thrity-Seventh Asilomar Conference on Signals, Systems & Computers, 2003 · doi:10.1109/acssc.2003.1292216 · Cited by 6,253
Self-citationNo-Reference Image Quality Assessment in the Spatial Domain
Mittal, A., Moorthy, A. K., & Bovik, A. C. (2012) · IEEE Transactions on Image Processing · doi:10.1109/tip.2012.2214050 · Cited by 6,064 · cited 7×
“We also avoid using duplicate features in different models, such as the BRISQUE-like features in HIGRADE, and the C-DIIVINE features in V-BLIINDS.”
MethodInfluentialSelf-citationDeep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1512.03385 · Cited by 4,196
Image information and visual quality
Sheikh, H. R., & Bovik, A. C. (2006) · IEEE Transactions on Image Processing · doi:10.1109/tip.2005.859378 · Cited by 4,104 · cited 1×
“These natural scene statistics (NSS) models predictably deviate in the presence of distortions, thereby characterizing perceived quality degradations [47].”
BackgroundSelf-citation
Research works printed but not linked: 1
- [19] L. V. D. Maaten and G. Hinton, 'Visualizing data using t-SNE,' J. Mach. Learn. Research , vol. 9, no. Nov, pp. 2579-2605, 2008. -
Cited by
VBench: Comprehensive Benchmark Suite for Video Generative Models
Huang, Z., He, Y., Yu, J., Zhang, F., Si, C., Jiang, Y., Zhang, Y., Wu, T., Jin, Q., Chanpaisit, N., et al. (2024) · 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52733.2024.02060 · Cited by 407
“Meanwhile, the Video Quality Assessment (VQA) methods [63, 94, 108–114] are primarily designed for real videos, thereby neglecting the unique challenges posed by generative models, such as artifacts in synthesized videos.”
MethodSelf-citationFAST-VQA: Efficient End-to-End Video Quality Assessment with Fragment Sampling
Wu, H., Chen, C., Hou, J., Liao, L., Wang, A., Sun, W., Yan, Q., & Lin, W. (2022) · Lecture Notes in Computer Science · doi:10.1007/978-3-031-20068-7_31 · Cited by 244
Perceptual video quality assessment: a survey
Min, X., Duan, H., Sun, W., Zhu, Y., & Zhai, G. (2024) · Science China Information Sciences · doi:10.1007/s11432-024-4133-3 · Cited by 218
RAPIQUE: Rapid and Accurate Video Quality Prediction of User Generated Content
Tu, Z., Yu, X., Wang, Y., Birkbeck, N., Adsumilli, B., & Bovik, A. C. (2021) · IEEE Open Journal of Signal Processing · doi:10.1109/ojsp.2021.3090333 · Cited by 215
“Among these, BRISQUE [8], GM-LOG [11], FRIQUEE [12], V-BLIINDS [9], and VIDEVAL [6] have leveraged different sets of natu-ral scene statistics (NSS)-based quality-aware features, using them to train shallow regressors to predict subjective quality scores.”
BackgroundMethodInfluentialSelf-citationWu, H., Zhang, E., Liao, L., Chen, C., Hou, J., Wang, A., Sun, W., Yan, Q., & Lin, W. (2023) · 2023 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01843 · Cited by 215
“Recent Video Quality Assessment (VQA) studies have gathered enormous human quality opinions [1–5] on in-the-wild user-generated contents (UGC) and attempted to use machine algorithms [6–8] to learn and predict these opinions, known as the UGCVQA problem [9].”
BackgroundInfluentialBlindly Assess Quality of In-the-Wild Videos via Quality-Aware Pre-Training and Motion Perception
Li, B., Zhang, W., Tian, M., Zhai, G., & Wang, X. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3164467 · Cited by 171
“5537 [73] to partition CombinedU into 1,469 low quality and 1,458 high quality materials, and tabulate the comparisons in the “Quality” column of Table VI.”
BackgroundMethodInfluentialLarge language models in law: A survey
Lai, J., Gan, W., Wu, J., Qi, Z., & Yu, P. S. (2024) · AI Open · doi:10.1016/j.aiopen.2024.09.002 · Cited by 362
Neighbourhood Representative Sampling for Efficient End-to-End Video Quality Assessment
Wu, H., Chen, C., Liao, L., Hou, J., Sun, W., Yan, Q., Gu, J., & Lin, W. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3319332 · Cited by 145
“In recent years, several methods [3], [4], [37] choose to first handcraft quality-sensitive features and then regress them to subjective mean opinion scores (MOS), in order to better fit the human perception.”
BackgroundMethodInfluentialQuality-aware Pretrained Models for Blind Image Quality Assessment
Zhao, K., Yuan, K., Sun, M. B., Li, M., & Wen, X. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.02136 · Cited by 163
“Due to the fact that the quality of an image is related to various complicated factors [65,71], a large number of images are needed to ensure the diversity of negative samples [7].”
BackgroundDisCoVQA: Temporal Distortion-Content Transformers for Video Quality Assessment
Wu, H., Chen, C., Liao, L., Hou, J., Sun, W., Yan, Q., & Lin, W. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3249741 · Cited by 111
“Classical VQA methods design on spatial [11], [27], [30] or spatial-temporal [20], [28], [32], [38] handcrafted features and regress on these features for quality modeling.”
BackgroundMethodResultInfluentialAn AIGC-empowered methodology to product color matching design
Wu, F., Hsiao, S., & Lu, P. (2023) · Displays · doi:10.1016/j.displa.2023.102623 · Cited by 100
A Comparative Evaluation Of Temporal Pooling Methods For Blind Video Quality Assessment
Tu, Z., Chen, C.-J., Chen, L.-H., Birkbeck, N., Adsumilli, B., & Bovik, A. C. (2020) · 2020 IEEE International Conference on Image Processing (ICIP) · doi:10.1109/icip40778.2020.9191169 · Cited by 71
“Simple temporal average pooling is a widely used scheme to augment BVQA models [5,7,10].”
MethodSelf-citationEvaluating Point Cloud From Moving Camera Videos: A No-Reference Metric
Zhang, Z., Sun, W., Zhu, Y., Min, X., Wu, W., Chen, Y., & Zhai, G. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3340894 · Cited by 93
Sun, W., Wang, T., Min, X., Yi, F., & Zhai, G. (2021) · IEEE International Conference on Multimedia & Expo Workshops (ICMEW) · doi:10.1109/icmew53276.2021.9455999 · Cited by 88
“VIDEVAL [12] further combines the selected features from typical NR I/VQA methods to train a SVR model to regress them into the video quality.”
InfluentialContrastive Self-Supervised Pre-Training for Video Quality Assessment
Chen, P., Li, L., Wu, J., Dong, W., & Shi, G. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3130536 · Cited by 62
Analysis of Video Quality Datasets via Design of Minimalistic Video Quality Models
Sun, W., Wen, W., Min, X., Lan, L., Zhai, G., & Ma, K. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3385364 · Cited by 72
“Our models consist of four basic building blocks: a video preprocessor, a spatial quality analyzer, an optional temporal quality analyzer, and a quality regressor.”
Study of the Subjective and Objective Quality of High Motion Live Streaming Videos
Shang, Z., Ebenezer, J. P., Wu, Y., Wei, H., Sethuraman, S., & Bovik, A. C. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3136723 · Cited by 54
“Frame based algorithms can be applied at a lower frame rate with little loss of prediction efficacy [62].”
BackgroundSelf-citationShen, W., Zhou, M., Liao, X., Jia, W., Xiang, T., Fang, B., & Shang, Z. (2022) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2022.3164332 · Cited by 49
“Tu et al. [29] created a fusion-based model that extracts 60 statistical features by employing a feature selection strategy, which can balance the trade-off between VQA performance and efficiency.”
BackgroundMethodKVQ: Kwai Video Quality Assessment for Short-form Videos
Lu, Y., Li, X., Pei, Y., Yuan, K., Xie, Q., Qu, Y., Sun, M. W., Zhou, C., & Chen, Z. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.02453 · Cited by 90
“In recent years, to develop more realistic and challenging video quality assessment (VQA) for user-generated content (UGC), many UGC databases [11, 37, 44, 47, 56, 67, 74], have been established collecting videos with authentic distortions.”
BackgroundMethodSpatiotemporal Representation Learning for Blind Video Quality Assessment
Liu, Y., Wu, J., Li, L., Dong, W., Zhang, J., & Shi, G. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3114509 · Cited by 48
Qu, B., Liang, X., Sun, S., & Gao, W. (2024) · arXiv · doi:10.48550/arxiv.2404.13573 · Cited by 44
A Subjective and Objective Study of Space-Time Subsampled Video Quality
Lee, D. Y., Paul, S., Bampis, C. G., Ko, H., Kim, J.-H., Jeong, S., Homan, B., & Bovik, A. C. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3137658 · Cited by 47
“The trained models include VMAF, VSTR, BRISQUE, TLVQM, and VIDEVAL [42] which each consists of 6, 16, 36, 75, and TABLE III PERFORMANCE COMPARISON OF VQA/IQA MODELS ON THE ETRI-LIVE STSVQ DATABASE ACROSS DIFFERENT SPATIAL SUBSAMPLING LEVELS.”
MethodSelf-citationSubjective and Objective Audio-Visual Quality Assessment for User Generated Content
Cao, Y., Min, X., Sun, W., & Zhai, G. (2023) · IEEE Transactions on Image Processing · doi:10.1109/tip.2023.3290528 · Cited by 49
“Tuet al. [33] propose that natural scene statistics (NSS) features are implicitly resolution-aware, while CNN features are not.”
BackgroundMethodQu, B., Liang, X., Sun, S., & Gao, W. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00659 · Cited by 61
“Classic NR-VQA meth-ods adopt handcrafted features as evaluation metrics [3, 15, 16, 29, 35, 40].”
Exploring the Effectiveness of Video Perceptual Representation in Blind Video Quality Assessment
Liao, L., Xu, K., Wu, H., Chen, C., Sun, W., Yan, Q., & Lin, W. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3547849 · Cited by 50
“One direct way is to compute frame-level quality scores and then express their relative importance over time by applying temporal pooling to these frame-level quality scores [43, 44, 57].”
BackgroundMethodInfluentialLearning Spatiotemporal Interactions for User-Generated Video Quality Assessment
Zhu, H., Chen, B., Zhu, L., & Wang, S. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3207148 · Cited by 42
“Generally speaking, UGC-VQA can be categorized into subjective and objective quality assessment [3].”
BackgroundMethodInfluentialCONVIQT: Contrastive Video Quality Estimator
Madhusudana, P. C., Birkbeck, N., Wang, Y., Adsumilli, B., & Bovik, A. C. (2023) · IEEE Transactions on Image Processing · doi:10.1109/tip.2023.3310344 · Cited by 56
“VIDEVAL [37] uses an ensemble approach whereby features from multiple NR models were carefully selected, yielding a feature set that was shown to perform well on UGC videos.”
InfluentialSelf-citationUnsupervised Curriculum Domain Adaptation for No-Reference Video Quality Assessment
Chen, P., Li, L., Wu, J., Dong, W., & Shi, G. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00513 · Cited by 53
“In [39], Tu et al. proposed a fusion-based VQA model called VIDEVAL, using a feature ensemble and selection procedure on top of existing efficient NR-VQA models.”
MethodHe, Y., Liu, D., Guo, R., & Siping, G. (2023) · Psychology Research and Behavior Management · doi:10.2147/prbm.s415832 · Cited by 43
EarthVQANet: Multi-task visual question answering for remote sensing image understanding
Wang, J., Ma, A., Chen, Z., Zheng, Z., Wan, Y., Zhang, L., & Zhong, Y. (2024) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2024.05.001 · Cited by 50
Subjective and Objective Quality of Experience of Free Viewpoint Videos
Yan, J., Li, J., Fang, Y., Che, Z., Xia, X., & Liu, Y. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3177127 · Cited by 33
Tu, Z., Lin, J., Wang, Y., Adsumilli, B., & Bovik, A. C. (2020) · IEEE Signal Processing Letters · doi:10.1109/lsp.2020.3024985 · Cited by 31
Self-citationFineVQ: Fine-Grained User Generated Content Video Quality Assessment
Duan, H., Hu, Q., Wang, J., Yang, L., Xu, Z., Lü, L., Min, X., Cai, C., Ye, T., Zhang, X., et al. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.00305 · Cited by 93
“…image quality assessment (IQA) methods, including NIQE [39], QAC [69], HOSA [68]; (2) traditional video quality assessment (VQA) methods, including TLVQM [22], VIDEVAL [53], RAPIQUE [54]; (3) DNN-based VQA methods, including VSFA [24], GSTVQA [2], SimpleVQA [49], FAST-VQA [60], DOVER [62].”
Self-Supervised Representation Learning for Video Quality Assessment
Jiang, S., Sang, Q., Hu, Z., & Liu, L. (2022) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2022.3197904 · Cited by 31
AI-Empowered Persuasive Video Generation: A Survey
Liu, C., & Yu, H. (2023) · ACM Computing Surveys · doi:10.1145/3588764 · Cited by 50
“InceptionTime [33] is utilized as the regressor network, which is a CNN that is designed for time series regression.”
MethodCompletely Blind Quality Assessment of User Generated Video Content
Kancharla, P., & Channappayya, S. S. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3130541 · Cited by 43
A Survey on Optimizing Mobile Delivery of 360° Videos: Edge Caching and Multicasting
Mahmoud, M., Rizou, S., Panayides, A. S., Kantartzis, N. V., Karagiannidis, G. K., Lazaridis, P. I., & Zaharis, Z. D. (2023) · IEEE Access · doi:10.1109/access.2023.3292335 · Cited by 41
“No-reference (i.e., blind) methods that utilize regularized natural scene statistics descriptors, such as normalized pixel intensities distribution, or employ machine learning to learn from a set of distortions, are increasingly used and likewise are emerging DL models [70].”
BackgroundVDPVE: VQA Dataset for Perceptual Video Enhancement
Gao, Y., Cao, Y., Kou, T., Sun, W., Dong, Y., Liu, X., Min, X., & Zhai, G. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00152 · Cited by 36
“Then, we test the performance of several state-of-the-art VQA methods on the test set, including V-BLIINDS [23], TLVQM [11], VIDEVAL [29], RAPIQUE [30], FastVQA [37], VSFA [14], and BVQA [13].”
MethodCOVER: A Comprehensive Video Quality Evaluator
He, C., Zheng, Q., Zhu, R., Zeng, X., Fan, Y., & Tu, Z. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00589 · Cited by 38
“Based on the optimal weights from stage 2, we add two SCGBs to the model.”
BackgroundMethodSelf-citationQuality Assessment of UGC Videos Based on Decomposition and Recomposition
Liu, Y., Wu, J., Li, L., Dong, W., & Shi, G. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3209007 · Cited by 32
“Eight UGC-VQA methods (i.e., vBLIINDS [24], VSFA [16], TL-VQM [11], CNN-TLVQM [12], NSTSS [29], VIDEVAL [13], RAPIQUE [14], and CoINVQ [32]) are also adopted for the performance comparison.”
BackgroundMethodInfluentialZoom-VQA: Patches, Frames and Clips Integration for Video Quality Assessment
Zhao, K., Yuan, K., Sun, M. B., & Wen, X. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00137 · Cited by 44
“…and representative features, including advanced backbones and modules [21, 49], quality-aware pre-training [30, 61] and multiple hand-crafted features ensemble [12, 45], and 2) designing quality score prediction head, such as normal regression heads [40, 45], patches fusion [59] and IP-NLR [49].”
BackgroundMethodA Completely Blind Video Quality Evaluator
Zheng, Q., Tu, Z., Zeng, X., Bovik, A. C., & Fan, Y. (2022) · IEEE Signal Processing Letters · doi:10.1109/lsp.2022.3215311 · Cited by 36
“It is also worth mentioning that VIQE delivered performance comparable to that of OA BIQA/BVQA models (trained onMOSlabels)ontheLIVE-VQC[29]andYouTube-UGC[31], andevensurpassedsomeleadinglearning-basedmodelson KoNViD-1k[30].”
BackgroundMethodInfluentialSelf-citationExploring Opinion-Unaware Video Quality Assessment with Semantic Affinity Criterion
Wu, H., Liao, L., Hou, J., Chen, C., Zhang, E., Wang, A., Sun, W., Yan, Q., & Lin, W. (2023) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme55011.2023.00070 · Cited by 37
StableVQA: A Deep No-Reference Quality Assessment Model for Video Stability
Kou, T., Liu, X., Sun, W., Jia, J., Min, X., Zhai, G., & Liu, N. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3611860 · Cited by 37
“[18, 20, 24, 33] utilize NR Image Quality Assessment (IQA) models on frames of videos and pool the results as the video quality score, while [11, 19, 27] utilize tailored handcrafted features for assessing video quality.”
MethodLMM-VQA: Advancing Video Quality Assessment With Large Multimodal Models
Ge, Q., Sun, W., Zhang, Y., Li, Y., Ji, Z., Sun, F., Jui, S., Min, X., & Zhai, G. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3571788 · Cited by 59
“Knowledge-driven BVQA models primarily utilize hand-crafted features derived through prior knowledge of visual quality perception to evaluate video quality [21, 39, 58, 59].”
BackgroundMethodInfluentialDDH-QA: A Dynamic Digital Humans Quality Assessment Database
Zhang, Z., Zhou, Y., Sun, W., Lu, W. J., Min, X., Wang, Y., & Zhai, G. (2023) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme55011.2023.00429 · Cited by 28
Self-citationA Spatial–Temporal Video Quality Assessment Method via Comprehensive HVS Simulation
Zhang, A., Wang, Y., Tang, W., Li, L., & Kwong, S. (2023) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2023.3338615 · Cited by 36
“Our HVS-5M is compared with 25 leading methods, including VQA methods, such as VIIDEO [66], V-BLIINDS [18], TLVQM [67], VIDEVAL [53], CNN-TLVQM [54], VSFA [36], GSTVQA [68], MDTVSFA [69], RAPIQUE [70], PVQ [52], BVQA-2022 [71], CONVIQT [74], and IQA methods, including NIQE [55], IL-NIQE [56],…”
MethodSelf-citationCompletely Blind Quality Assessment of User Generated Video Content
Kancharla, P., & Channappayya, S. S. (2022) · Research Archive of Indian Institute of Technology Hyderabad (Indian Institute of Technology Hyderabad) · Cited by 23
Starvqa: Space-Time Attention for Video Quality Assessment
Xing, F., Wang, Y., Wang, H., Li, L., & Zhu, G. (2022) · Proceedings - International Conference on Image Processing · doi:10.1109/icip46576.2022.9897881 · Cited by 38
“In this part, we compare the proposed StarVQA with five state-ofthe-art methods, including BRISQUE [26], TLVQM [27], VIDEVAL [28], VSFA [7], and PVQ [10].”
MethodSelf-citationLight-VQA: A Multi-Dimensional Quality Assessment Model for Low-Light Video Enhancement
Dong, Y., Liu, X., Gao, Y., Zhou, X., Tan, T., & Zhai, G. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3611923 · Cited by 29
“VIDEVAL [35] combines existing VQA methods together and proposes a feature selection strategy, which can choose appropriate features and then fuse them efficiently to predict the quality scores of videos.”
InfluentialStudy of Spatio-Temporal Modeling in Video Quality Assessment
Fang, Y., Li, Z., Yan, J., Sui, X., & Liu, H. (2023) · IEEE Transactions on Image Processing · doi:10.1109/tip.2023.3272480 · Cited by 23
“Tu et al. [47] proposed a fusion-based model called VIDEVAL, which employs a feature selection strategy and a support vector regressor (SVR) as the regression model to learn the feature-to-score mapping.”
MethodWu, T., Jiang, N., Kumar, T. B. J., & Chen, M. (2024) · Cogent Business & Management · doi:10.1080/23311975.2024.2329247 · Cited by 60
“The concept of perceived variety refers to users’ subjective perception of the platform’s content diversity (Tu et al., 2021).”
Subjective and Objective Analysis of Streamed Gaming Videos
Yu, X., Ying, Z., Birkbeck, N., Wang, Y., Adsumilli, B., & Bovik, A. C. (2023) · IEEE Transactions on Games · doi:10.1109/tg.2023.3293093 · Cited by 32
“Finally, we summarize this article and discuss possible future work in Section VII.”
BackgroundMethodSelf-citationNTIRE 2023 Quality Assessment of Video Enhancement Challenge
Liu, X., Min, X., Sun, W., Zhang, Y., Zhang, K., Timofte, R., Zhai, G., Gao, Y., Cao, Y., Kou, T., et al. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00158 · Cited by 26
“We compare the performance of submitted methods with several state-of-the-art NR VQA methods on the testing set, including V-BLIINDS [56], TLVQM [31], VIDEVAL [69], RAPIQUE [70], FastVQA [80], VSFA [33], BVQA [45], and SimpleVQA [65].”
BackgroundMethodInfluentialWang, J., Duan, H., Zhai, G., Wang, J., & Min, X. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.01758 · Cited by 64
No-Reference Video Quality Assessment Using Local Structural and Quality-Aware Deep Features
Vishwakarma, A. K., & Bhurchandi, K. M. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3273654 · Cited by 23
“Convolutional neural network (CNN)-based VSFA [13] yielded comparable performance with VIDEVAL [4], but lagged considerably (not in the top two) compared to TLVQM [12] in the presence of 1557-9662 © 2023 IEEE.”
BackgroundMethodResultInfluentialRankDVQA: Deep VQA based on Ranking-inspired Hybrid Training
Feng, C., Danier, D., Zhang, F., & Bull, D. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00167 · Cited by 28
“To benchmark the performance of the proposed methods, we tested ten full reference quality assessment methods, including three conventional ones: PSNR, SSIM [63], and MS-SSIM [65]; and five deep VQA methods:4 WaDIQA [4], DeepQA [21], DeepVQA [22], C3DVQA [69] and DISTS [10]; and two machine learning (SVM regression) b…”
MethodWhy no-reference metrics for image and video quality lack accuracy and reproducibility
Pinson, M. H. (2023) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2022.3191059 · Cited by 19
“Similarly, [48] uses three datasets ( KoNViD-1K , LIVE-Qualcomm, and CVD2014 ) and [98] uses three datasets ( KoNViD-1K , LIVE-VQC and YouTube-UGC ), which they refer to as UGC-VQA .”
MethodBlind VQA on 360° Video via Progressively Learning From Pixels, Frames, and Video
Yang, L., Xu, M., Li, S., Guo, Y., & Wang, Z. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3226417 · Cited by 26
“2D videos [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], comparatively few efforts have been devoted to VQA for 360 videos [19], [20], [21], [22], [23], [24], [25], [26], [27].”
BackgroundMethodInfluentialA Perceptual Quality Metric for Video Frame Interpolation
Hou, Q., Ghildyal, A., & Liu, F. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19784-0_14 · Cited by 37
GAMIVAL: Video Quality Prediction on Mobile Cloud Gaming Content
Chen, Y.-C., Saha, A., Davis, C., Qiu, B., Wang, X., Gowda, R., Katsavounidis, I. X., & Bovik, A. C. (2023) · IEEE Signal Processing Letters · doi:10.1109/lsp.2023.3255011 · Cited by 19
“Supervised VQA methods like BRISQUE, TLVQM, VIDEVAL, RAPIQUE, and GAME-VQP operate by training an SVR to learn feature-to-score mappings.”
BackgroundMethodSelf-citationStudy of Subjective and Objective Quality Assessment of Mobile Cloud Gaming Videos
Saha, A., Chen, Y.-C., Davis, C., Qiu, B., Wang, X., Gowda, R., Katsavounidis, I. X., & Bovik, A. C. (2023) · IEEE Transactions on Image Processing · doi:10.1109/tip.2023.3281170 · Cited by 27
“The algorithms VSFA and NDNet-Gaming were implemented in Python, GAMIVAL was implemented partly in MATLAB and partly in Python, while all the other algorithms were implemented in MATLAB.”
MethodInfluentialSelf-citationA no-reference quality assessment metric for dynamic 3D digital human
Chen, S., Zhang, Z., Zhou, Y., Sun, W., & Min, X. (2023) · Displays · doi:10.1016/j.displa.2023.102540 · Cited by 24
2BiVQA: Double Bi-LSTM-based Video Quality Assessment of UGC Videos
Telili, A., Fezza, S. A., Hamidouche, W., & Meftah, H. F. Z. B. (2023) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3632178 · Cited by 25
“…Image Quality Assessment (BIQA)/Blind Video Quality Assessment (BVQA) methods achieve good performance on synthetic distortion datasets [3], their performance on UGC videos remains far from satisfactory [4]–[8], and predicting the quality of UGC videos is still challenging and unsolved problem.”
BackgroundMethodInfluentialSubjective Quality Assessment of User-Generated Content Gaming Videos
Yu, X., Tu, Z., Ying, Z., Bovik, A. C., Birkbeck, N., Wang, Y., & Adsumilli, B. (2022) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw54805.2022.00013 · Cited by 31
“When comparing the distribution of SROCC values, VIDEVAL and RAPIQUE behave statistically the same.”
MethodInfluentialSelf-citationQ-Boost: On Visual Quality Assessment Ability of Low-Level Multi-Modality Foundation Models
Zhang, Z., Wu, H., Ji, Z., Li, C., Zhang, E., Sun, W., Liu, X., Min, X., Sun, F., Jui, S., et al. (2024) · IEEE International Conference on Multimedia and Expo workshops · doi:10.1109/icmew63481.2024.10645451 · Cited by 39
“• The opinion-aware VQA methods include TLVQM [25], VSFA [7], and VIDEVAL [26].”
MethodCLIPVQA: Video Quality Assessment via CLIP
Xing, F., Li, M., Wang, Y., Zhu, G., & Cao, X. (2024) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2024.3511927 · Cited by 30
“Tu et al. [47] introduced a fusion-based VQA model involving a feature ensemble and selection procedure.”
Self-citationBenchmarking Multi-dimensional AIGC Video Quality Assessment: A Dataset and Unified Model
Zhang, Z., Sun, W., Li, X., Jia, J., Min, X., Zhang, Z., Li, C., Chen, Z., Wang, P., Sun, F., et al. (2025) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3749844 · Cited by 62
Thqa: A Perceptual Quality Assessment Database for Talking Heads
Zhou, Y., Zhang, Z., Sun, W., Liu, X., Min, X., Wang, Z., Zhang, X.-P. S., & Zhai, G. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10647507 · Cited by 28
Study of Subjective and Objective Naturalness Assessment of AI-Generated Images
Chen, Z., Sun, W., Wu, H., Zhang, Z., Jia, J., Huang, R., Min, X., Zhai, G., & Zhang, W. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3509032 · Cited by 71
“Besides, we follow the procedure in [64] to quantify the relative range and coverage uniformity of these datasets over each feature space, which measure the inter-and intra-dataset differences, respectively.”
A neural network based framework for effective laparoscopic video quality assessment
Khan, Z. A., Beghdadi, A., Kaaniche, M., Cheikh, F. A., & Gharbi, O. (2022) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2022.102121 · Cited by 23
“Before providing an overview of the VQA methods, let us emphasize here that most of the developed works have been devoted to natural video data [10]–[14] and only few works have”
MethodEnd-to-End Blind Video Quality Assessment Based on Visual and Memory Attention Modeling
Guan, X., Li, F., Zhang, Y., & Cosman, P. C. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3189251 · Cited by 19
FAVER: Blind quality prediction of variable frame rate videos
Zheng, Q., Tu, Z., Madhusudana, P. C., Zeng, X., Bovik, A. C., & Fan, Y. (2024) · Signal Processing Image Communication · doi:10.1016/j.image.2024.117101 · Cited by 28
“Inspired by the success of VMAF [3], VIDEVAL [4] uses a similar ensemble model that fuses features taken from several successful BVQA models, based on a feature selection procedure.”
BackgroundMethodInfluentialSelf-citationSubjective and Objective Quality Assessment of Colonoscopy Videos
Yue, G., Zhang, L., Du, J., Zhou, T., Zhou, W., & Lin, W. (2024) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2024.3461737 · Cited by 25
“4) Compared Methods: We compare our DARNet with nine state-of-the-art VQA methods, including TLVQM [39], VIDEVAL [40], VSFA [25], Simple-VQA [30], BVQA [31], FAST-VQA-M [27], FASTER-VQA [41], Light-VQA [51], and DOVER [15].”
MoA-VR: A Mixture-of-Agents System Toward All-in-One Video Restoration
Liu, L., Cai, C., Shen, S., Liang, J., Ouyang, W., Ye, T., Mao, J. Z., Duan, H., Yao, J., Zhang, X., et al. (2025) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2025.3627128 · Cited by 35
“The resulting degradation profile serves as the initial condition for subsequent routing decisions.”
Perceptual video quality assessment: the journey continues!
Saha, A., Pentapati, S. K., Shang, Z., Pahwa, R., Chen, B., Gedik, H. E., Mishra, S., & Bovik, A. C. (2023) · Frontiers in Signal Processing · doi:10.3389/frsip.2023.1193523 · Cited by 21
“Similar to VMAF in FR-VQA, VIDEVAL (Tu et al., 2021a) is an NR-VQA model developed to evaluate the quality of UGC videos by combining subsets of statistical features extracted from various existing NR-IQA and NR-VQA models.”
MethodSelf-citationA Blind Video Quality Assessment Method via Spatiotemporal Pyramid Attention
Shen, W., Zhou, M., Wei, X., Wang, H., Fang, B., Ji, C., Zhuang, X., Wang, J. M., Luo, J. T., Pu, H., et al. (2023) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2023.3340031 · Cited by 19
“In this section, we conduct a cross-database experimental scheme to verify the robustness of our method compared to other methods (VBLIINDS [48], TLVQM [17], VSFA [8] and VIDEVAL [15]).”
BackgroundMethodOmnidirectional Video Quality Assessment With Causal Intervention
Hu, Z., Liu, L., & Sang, Q. (2024) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2023.3342707 · Cited by 19
“…as motion features [12], [15], natural scene statistical (NSS) features [12], [18], [19], [20], gradient features [13], [21], [22], and distortion-aware features [13], [15], [23] from videos and trained support vector regressors (SVR) or random forests to map the features to quality predictions.”
BackgroundMethodInfluentialSecurity and Privacy Challenges of AIGC in Metaverse: A Comprehensive Survey
Zhang, S., Li, H., Sun, K., Chen, H., Wang, Y., & Li, S. (2025) · ACM Computing Surveys · doi:10.1145/3729419 · Cited by 19
“AIGC technology can provide highly realistic 3D appearance representations for visual natural interactions, versatile dialogue generation, and lexible personalized interaction response generation.”
Self-citationVBench++: Comprehensive and Versatile Benchmark Suite for Video Generative Models
Huang, Z., Zhang, F., Xu, X., He, Y., Yu, J., Dong, Z., Ma, Q., Chanpaisit, N., Si, C., Jiang, Y., et al. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3633890 · Cited by 351
“Meanwhile, the VideoQualityAssessment (VQA) methods[46], [47], [48], [49], [50], [51], [52], [53], [54] are primarily designed for real videos, thereby neglecting the unique challenges posed by generative models, such as artifacts in synthesized videos.”
Self-citationTowards adversarial robustness verification of no-reference image- and video-quality metrics
Shumitskaya, E., Antsiferova, A., & Vatolin, D. S. (2023) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2023.103913 · Cited by 25
AIS 2024 Challenge on Video Quality Assessment of User-Generated Content: Methods and Results
Conde, M. V., Zadtootaghaj, S., Barman, N., Timofte, R., He, C., Zheng, Q., Zhu, R., Tu, Z., Wang, H., Chen, X., et al. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00591 · Cited by 19
Self-citationSubjective and Objective Quality-of-Experience Assessment for 3D Talking Heads
Zhou, Y., Zhang, Z., Sun, W., Liu, X., Min, X., & Zhai, G. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3680964 · Cited by 15
Making Video Quality Assessment Models Robust to Bit Depth
Ebenezer, J. P., Shang, Z., Wu, Y., Wei, H., Sethuraman, S., & Bovik, A. C. (2023) · IEEE Signal Processing Letters · doi:10.1109/lsp.2023.3268602 · Cited by 12
Self-citationGeometry-Aware Video Quality Assessment for Dynamic Digital Human
Zhang, Z., Zhou, Y., Sun, W., Min, X., & Zhai, G. (2023) · Proceedings - International Conference on Image Processing · doi:10.1109/icip49359.2023.10222061 · Cited by 16
“To incorporate spatial and temporal features, handcrafted-based methods have been proposed, such as VIIDEO [5], V-BLIINDS [6], TLVQM [7], and VIDEVAL [8].”
MethodDeep Learning Approach for No-Reference Screen Content Video Quality Assessment
Kwong, N.-W., Chan, Y., Tsang, S., Huang, Z., & Lam, K. (2024) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2024.3374042 · Cited by 12
“…[6], GMSD [7], BRISQUE [8], and NIQE [9]), four SCIQA methods (ESIM [21], SQMS [25], SVQI [22] and Yang’s work [20]), five NSVQA methods (MOVIE [11], STMAD [12], STRRED [13], VSFA [14] and VIDEVAL [16]) and four SCVQA methods (SGFTM [28], MS-RSDS [29], HSFM [30] and Li’s work [31]), were included.”
MethodLiu, B. (2024) · Scientific Reports · doi:10.1038/s41598-024-72343-w · Cited by 15
Socialized Learning: A Survey of the Paradigm Shift for Edge Intelligence in Networked Systems
Wang, X., Zhao, Y., Qiu, C., Hu, Q., & Leung, V. C. M. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3482978 · Cited by 17
“Integrating SL into AIGC enhances its ability to learn user interactions and collective insights, personalizing content, but also unique challenges beyond traditional approaches [268].”
Quality Feature Learning via Multi-Channel CNN and GRU for No-Reference Video Quality Assessment
Kwong, N.-W., Chan, Y., Tsang, S., & Lun, D. P.-K. (2023) · IEEE Access · doi:10.1109/access.2023.3259101 · Cited by 12
“Twelve NR-VQA methods, BRISQUE [17], NIQE [15], V-BLIINDS [21], VIIDEO [20], TLVQM [10], VSFA [12], CNN-TLVQM [13], HEKE [24], RAPIQUE [23], VIDEVAL [71], MDTVSFA [52], and LSCT-PHIQNet [30] were included.”
MethodQuality Assessment for AI Generated Images With Instruction Tuning
Wang, J., Duan, H., Zhai, G., & Min, X. (2026) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2026.3651009 · Cited by 86
“Numerous image and video quality assessment models have been proposed in the literature [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], which aims to objectively and quantitatively evaluate the perceptual quality.”
Panchal, H. D., & Shah, H. B. (2023) · Multimedia Systems · doi:10.1007/s00530-023-01123-9 · Cited by 12
“The proposed algorithm employs a total 21 video quality assessment attributes or features [6, 22–29] tabulated in Table 2.”
BackgroundMethodInfluentialDVL2021: An ultra high definition video dataset for perceptual quality study
Xing, F., Wang, Y., Wang, H., He, J., & Yuan, J. (2021) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2021.103374 · Cited by 12
Self-citationEnhancing blind video quality assessment with rich quality-aware features
Sun, W., Cao, L., Jia, J., Zhang, Z., Zhang, Z., Min, X., & Zhai, G. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.130452 · Cited by 40
“Knowledge-driven BVQA models [5,12,29,35,43,44] utilize carefully designed handcrafted features to quantify the video quality.”
InfluentialWu, W., Hu, S., Xiao, P., Deng, S., Li, J., Chen, Y., & Li, K. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00183 · Cited by 13
“In order to conduct a comprehensive assessment of the proposed method’s performance, we compare it with several popularly quality assessment models, namely BRISQUE [26], TLVQM [17], VIDEVAL [36], RAPIQUE [37], VSFA [20], PVQ [44], BVQA [19], and SimpleVQA [35].”
BackgroundMethodInfluentialBVI-VFI: A Video Quality Database for Video Frame Interpolation
Danier, D., Zhang, F., & Bull, D. (2023) · IEEE Transactions on Image Processing · doi:10.1109/tip.2023.3327912 · Cited by 19
“These include image models, NIQE [116], BRISQUE [117] and deepIQA-NR [113], and video models, VIIDEO [118], VBLIINDS [119], VIDEVAL [120], RAPIQUE [121], ChipQA [122], TLVQM [123], FastVQA [124] and FAVER [78].”
Method360° video quality assessment based on saliency-guided viewport extraction
Yang, F., Yang, C., An, P., & Huang, X. (2024) · Multimedia Systems · doi:10.1007/s00530-024-01285-0 · Cited by 12
“Tu et al. [12] proposed the VIDEVAL algorithm to construct an initial feature set by selecting features from existing best-performing VQA models, which use machine learning models to learn important features.”
HDR or SDR? A Subjective and Objective Study of Scaled and Compressed Videos
Ebenezer, J. P., Shang, Z., Chen, Y., Wu, Y., Wei, H., Sethuraman, S., & Bovik, A. C. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3404890 · Cited by 16
“BRISQUE [27], VBLIINDS [28], VIDEVAL [29], RAPIQUE [30], ChipQA [31], HDR ChipQA [32] and NIQE [33] are NR video quality metrics that rely on neurostatistical models of visual perception.”
BackgroundSelf-citationVideo Quality Assessment of User Generated Content: A Benchmark Study and a New Model
Tu, Z., Chen, C.-J., Wang, Y., Birkbeck, N., Adsumilli, B., & Bovik, A. C. (2021) · Proceedings - International Conference on Image Processing · doi:10.1109/icip42928.2021.9506189 · Cited by 14
“It has been observed in [15] that LIVE-VQC videos generally contain more (camera) motions than KoNViD-1k and YouTube-UGC, and TLVQM computes multiple motion relevant features.”
Self-citationAudio-Visual Quality Assessment for User Generated Content: Database and Method
Cao, Y., Min, X., Sun, W., Zhang, X.-P. S., & Zhai, G. (2023) · Proceedings - International Conference on Image Processing · doi:10.1109/icip49359.2023.10222476 · Cited by 10
“We utilize the following handcrafted video and audio feature extractors: • Video: BRISQUE [12], CORNIA [13], V-BLIINDS [14], TLVQM [15], VIDEVAL [16], and RAPIQUE [17].”
MethodNo-Reference Video Quality Assessment Using Transformers and Attention Recurrent Networks
Kossi, K. K., Coulombe, S., & Desrosiers, C. (2024) · IEEE Access · doi:10.1109/access.2024.3468336 · Cited by 10
“Recent popular handcrafted methods include ChipQA [16], TLVQM [17] and VIDEVAL [18].”
MethodResult
Metadata
- Authors
- Tu, Z., Wang, Y., Birkbeck, N., Adsumilli, B., & Bovik, A. C.
- Affiliations
- The University of Texas at Austin (US)
- Google (United States) (US)
- Venue
- IEEE Transactions on Image Processing
- Published
- 2021
- Volume
- 30
- Pages
- 4449-4464
- ISSN
- 1057-7149 (ISSN-L)
- 1941-0042
- OpenAlex
- W3030701471
- Semantic Scholar
- a0d2b46a4ee9eca6f32df1c6f6bf7b1fbe6c0e0e
- Cited by
- 363
- References
- 81
- Citations by year
- 2026: 40 · 2025: 75 · 2024: 84 · 2023: 74 · 2022: 46 · 2021: 36
- FWCI
- 21.56
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Image Processing (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Video Quality Assessment
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image Processing Techniques
- Keywords
- blind video quality assessment, user-generated content, benchmark, feature selection, feature fusion, computational efficiency, perceptual video quality, in-the-wild videos, video streaming
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash and gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.